Review of FDA Warning Letters

The FDA provides an electronic reading room on its website that provides access to a great deal of useful information, including copies of Warning Letters issued by the FDA. Reviewing these letters can be useful because they provide insight into the inspection techniques and concerns of FDA inspectors. Additionally, they can be used to perform gap analyses of the processes used at your company in preparation for an FDA inspection. This article provides a summary of the review conducted for Warning Letters issued between 2000 and 2010 for products that are aseptically filled and for manufacturing processes identified as non-sterile.

Background

Employee training in a pharmaceutical company almost always includes instilling a 'healthy fear' of regulatory agency inspections. Most, if not all, countries have established governmental agencies that are responsible for the safety and efficacy of the medicines manufactured, sold and distributed within their country. These agencies typically have the legal authority to go into production facilities to assess various aspects of the manufacturing and distribution facility. As such, employees learn that the company's viability is dependent upon successfully passing these inspections.¹

The notification that a regulatory inspector is planning a visit to a company site can trigger concern regarding the expected outcome and frantic activities often take place in preparation. Since the implementation of the FDA's Systems Based inspection programme, however, it has become more difficult to successfully prepare for an inspection.

Originally, the Systems Based Inspection guide was introduced by the FDA as a pilot programme.² The method for evaluation entailed a portion of the FDA's regional offices using the programme for a 6-month trial period in 2001. The results were successful and the FDA officially implemented the programme in 2002 as the Compliance Program Guidance Manual for FDA Staff: Drug Manufacturing Inspections (Program 7356.002).²

Prior to the implementation of this guidance, companies prepared for inspections by thoroughly reviewing and assessing products with upcoming pre-approval inspections because these products tended to be the focus of the inspection. Under the new guidance, however, the FDA reviews companies in a systematic approach, which makes it more difficult to prepare. One valuable way to prepare for an inspection is to look at the FDA's inspection findings from other companies to determine which types of observations are also issues at your company.

For this study, the FDA's website was searched to review the Warning Letters issued between 2000 and mid-2010. The searches were limited to those referencing aseptic processing or non-sterile processing, and the data were summarised.^2,3

Warning Letter review

Data obtained from the Warning Letter review were analysed to determine the appropriate section of the GMPs, as stated in 21 CFR Part 211, that had been violated. This was then summarised to determine how many observations were made for each section of the GMPs. Following determination of the number of observations made, a ranking was performed to determine which section resulted in the most observations (rank = 1) and which resulted in the least (rank = 33 for aseptic; rank = 44 for non-sterile processing). Items with the same number of observations were given the same ranking within the type of processing used. These results are summarised in Table 1. The top ten categories of observations for each type of processing are summarised in Table 2.

Table 1: Number of FDA observations made per section of the Code of Federal Regulations, 21 CFR Part 211.

Although the product processing types were different (aseptic versus non-sterile), among the top ten categories of observations, several made the listing for both types of processes; for example: responsibilities of the quality control unit, equipment cleaning and maintenance, written procedures and deviations, general requirements (laboratory controls), testing and release for distribution, batch production and control records, and production record review.

Table 2: Top ten categories of FDA Warning Letter observations.

Overview of observations made by the FDA

The following sections describe some of the types of observations made by the FDA relative to the cited GMP section. Only sections that ranked in the top five for each process type are covered.

21 CFR §211.42

Design and construction features — Rank number 1 for aseptic

Examples of the types of observations made include:

• Material flow throughout the building is not designed to prevent contamination.

• Lack of separate or defined areas to prevent contamination or mix-ups.

• Failure to have an adequate air supply filtered through high efficiency particulate air filters under positive pressure, regardless of whether it is or is not laminar.

• Failure to have an adequate system for monitoring environmental conditions or the system established is inadequate. Air sampling is not performed at critical sites.

• Control systems used to prevent contamination do not include floors, walls, and ceilings of smooth, hard surfaces that are easily cleanable.

• Failure to establish an adequate system for cleaning and disinfecting the room and equipment.

• Studies were not performed to show the adequacy and efficacy of the cleaning and disinfection process used.

• Airflow pattern studies were not performed where connections were made.

• Inadequate containment control and monitoring programmes.

• Appropriate written procedures for prevention of microbiological contamination were not established or followed.

21 CFR §211.160

General requirements (laboratory controls) — Rank number 1 for non-sterile manufacturing and number 7 for aseptic processing

Examples of the types of observations made include:

• Failure to follow scientifically sound and appropriate sampling plans.

• Laboratory controls do not include establishment of scientifically sound and appropriate specifications.

• Laboratory equipment is not sufficiently calibrated and qualified at suitable intervals, in accordance with an established written programme.

• Laboratory controls are deficient. Procedures allow for averaging of out-of-specification (OOS) and within specification analytical test results from separate samples.

• New methods do not include sensitivity (limit of quantitation), linearity, accuracy or an appropriate precision method.

• Failure to qualify the supplier of the raw material.

• Failure to conduct OOS investigations.

• Methods used for stability testing are not stability indicating.

• Method validation is inadequate. It does not address all variables.

• Laboratory controls do not include determination of conformance to written specifications for the acceptance of each lot within a shipment of components used.

• Test devices are deficient. They did not meet established specifications and were used.

• Failure to ensure that complete data derived from all tests are documented at the time of performance.

• Analytical methods used are inadequate.

• Failure to follow your stability programme.

21 CFR §211.113

Control of microbiological contamination — Rank number 2 for aseptic processing

Examples of the types of observations made include:

• Failure to follow procedures designed to prevent microbiological contamination.

• Failure to adequately assess and monitor the aseptic environment.

• Failure to establish and follow appropriate written procedures.

• Procedures do not include validation of sterilisation processes.

• Failure to establish and maintain procedures to adequately control environmental conditions that could reasonably be expected to have an adverse effect on product quality.

• Failure to ensure that set-up personnel use aseptic techniques.

• Lack of appropriate written procedures.

• Environmental monitoring procedures are deficient. They do not include a scientific assessment to properly identify sampling sites.

21 CFR §211.100

Written Procedures; deviations — Rank number 2 for non-sterile processing and number 5 for aseptic processing

Examples of the types of observations made include:

• Written production and process control procedures not established.

• Written production and process control procedures not followed.

• Failure to adequately investigate, document, and explain the process deviation that occurred.

• Failure to appropriately validate the manufacturing processes currently used.

• Deviations were made without any documented justification.

• Failure of the quality control unit (QCU) to review and approve changes to established written procedures.

• Lack of process validation.

• Water systems are not validated.

• Failure to perform process validation.

• Failure to establish procedures for cleaning and maintenance of equipment.

21 CFR §211.192

Production record review — Rank number 3 for both types of processes

Examples of the types of observations made include:

• Failure to thoroughly investigate or maintain a written record of the investigation of any unexplained discrepancy or the failure of the batch or any of its components to meet any of its specifications.

• The QCU failed to review production records to assure that no errors have occurred, that they have been fully investigated, and conduct investigations.

• Failure to follow your procedure for investigations.

• OOS results are disregarded or negated without a documented investigation.

• Investigations were not performed in a timely manner.

• Investigations did not extend into other batches of the same product or other products that may have been affected.

• Incomplete investigation.

• Inadequate investigations.

• Disagreement with conclusion of investigation.

21 CFR §211.67

Equipment cleaning and maintenance — Rank number 4 for both types of processes

Examples of the types of observations made include:

• Failure to establish and follow adequate written procedures for equipment cleaning and maintenance.

• Failure to maintain records of cleaning of equipment, maintenance, sanitising and inspection.

• Deficient procedures.

• Inadequate validation studies.

• Equipment not cleaned at appropriate intervals.

• Equipment not properly maintained to prevent malfunctions.

• Cleaning validation studies did not look at levels of pharmaceutical or detergent residues. Methods used were not validated.

• Failure to establish the effectiveness of cleaning and disinfection procedures.

21 CFR §211.22

Responsibilities of the quality control unit — Rank number 5 for both processes

Examples of the types of observations made include:

• Failure to have an adequate quality assurance programme in place.

• Failure to establish and follow adequate written procedures applicable to the quality control unit (QCU).

• Failure to review production records to ensure no errors have occurred.

• Failure to have a QCU that has the authority and responsibility to approve or reject all components, in-process materials, drug products and all procedures or specifications.

• The QCU did not reject failing batches of drug product.

• The auditing staff failed to follow their procedure.

• The QCU failed to maintain authority and responsibility for conducting adequate investigations.

• OOS investigations did not determine a root cause for the deviation.

• Failure to have adequate laboratory testing facilities.

• All cGMP deficiencies were indicative of QCU's failure to meet requirements impacting the identity, strength, quality and purity of the drug products.

21 CFR §211.165

Testing and release for distribution — Rank number 5 for non-sterile processing and number 7 for aseptic processing

Examples of the types of observations made include:

• Failure to have, for each batch of drug product, appropriate laboratory determination of satisfactory conformance to final specifications.

• No assurance that all drug products meet USP requirements.

• Inadequate qualification of raw materials.

• Failure to reject a lot of drug product that did not meet all specifications.

• Failure to have an adequate validation procedure for computerised spreadsheets used for performing analytical calculations.

• Failure to have appropriate controls over computerised laboratory systems to ensure that changes or deletions of records are only performed by authorised personnel.

• Failure to establish criteria for sampling and testing.

21 CFR §211.166

Stability testing — Rank number 5 for non-sterile processes

Examples of the types of observations made include:

• Failure to have a written stability programme.

• Inadequate stability programme.

• Failure to have stability testing procedures.

• Failure to follow stability programme.

• No stability testing data developed.

• Failure to have data to support expiration dating.

Conclusions

In recent years, the number of Warning Letters issued for medicinal products regarding 21 CFR Part 211 has increased and there is a wealth of information available that can help other companies to avoid the same issues. In particular, several categories of observations were found to be significant for both aseptic and non-sterile processed pharmaceutical products, and can be traced back to the QCU. Whenever there are a significant number of deviations, it is also probable that you will obtain a statement such as: "The deficiencies described in the Form FDA 483 and this letter are indicative of your quality control unit not fulfilling its responsibility to assure the identity, strength, quality, and purity of your manufactured drug products."²

Reviewing the various Warning Letters indicates that for aseptic processes, typically senior, experienced investigators are conducting inspections. The investigators are very knowledgeable on US expectations for aseptic processing operations. In Warning Letters for both aseptic processes and non-sterile manufacturing, expectations seem to be the same regardless of the size of the manufacturing operation. They are also consistent whether the manufacturing site is within or outside of the US.²

Although GMPs have been in force for many years, there continues to be a large number of observations related to insufficient documentation and record keeping. Reviewing FDA Warning Letters is a useful tool in benchmarking your processes to current thinking.

Disclaimer

Every effort was made to accurately represent the Warning Letter information. Presence of the information in the Warning Letter is based upon the data available through the FDA Electronic Reading Room. It is known that companies may contest findings, and some may be subsequently changed or eliminated. Unfortunately, the information presented cannot show any actions that were not included on the government website.²

While titled 'Aseptic Processing' or 'Non-Sterile Processing', the Warning Letters were obtained by searching the FDA website for aseptic processing or non-sterile processing (including: searches for creams, lotions, tablets, syrups, ophthalmics, oral dosage forms and so forth) as a topic. As such, it is not possible to know from firsthand knowledge whether these sites only manufacture aseptically, terminally sterilise or manufacture only non-sterile products. Additionally, if errors (e.g., terminal sterilisation versus aseptic processing) have occurred, they are not intended. Observations identified in support of claims that products were misbranded or adulterated were not included unless they were specifically identified as GMP violations.^2,3

Jeanne Moldenhauer is Vice President of Excellent Pharma Consulting, Inc. 404 Washington Blvd, Mundelein, IL 60060 (USA). jeannemoldenhauer@yahoo.com Tel. +1 847 837 8191

References

1. J. Moldenhauer, Recent Warning Letters Review for Preparation of an Aseptic Processing Inspection. Volume 1 (PDA/DHI, Bethesda, Maryland, USA, 2010).

2. FDA, Compliance Program Manual, Document 7356.002: Drug Manufacturing Inspections (2002). www.fda.gov

3. J. Moldenhauer, Recent Warning Letters Review for Preparation of a Non-Sterile Processing Inspection. Volume 2 (PDA/DHI, Bethesda, Maryland, USA, 2010).